Long-Term Mortality and Health Status Outcomes Suzanne V. Arnold, MD, MHA, Kim G. Smolderen, PHD, Donna M. Buchanan, PHD, Yan Li, PHD John A. Spertus, MD, MPH Kansas City, Missouri Objectives This study sought to determine the association of chronic stress with long-term adverse outcomes after acute myocardial infarction (AMI). Background Chronic stress has been shown to be associated with the development of cardiovascular disease and, in the case of particular types of stress such as job and marital strain, with recurrent adverse events after AMI. Little is known, however, about the association of chronic stress with mortality and adverse health status outcomes in a general AMI population. Methods In a cohort of 4,204 AMI patients from 24 U.S. hospitals completing the Perceived Stress Scale-4 (sum scores ranging from 0 to 16) during hospitalization, moderate/high stress over the previous month was dened as scores in the top 2 quintiles (scores of 6 to 16). Detailed data on sociodemographics, psychosocial status, and clinical characteristics were collected at baseline. Outcomes included patients 1-year health status, assessed with the Seattle Angina Questionnaire, Short Form-12, and EuroQol Visual Analog Scale, and 2-year mortality. Results AMI patients with moderate/high stress had increased 2-year mortality compared with those having low levels of stress (12.9% vs. 8.6%; p 0.001). This association persisted after adjusting for sociodemographics, clinical factors (including depressive symptoms), revascularization status, and GRACE (Global Registry of Acute Coronary Events) discharge risk scores (hazard ratio: 1.42: 95% condence interval: 1.15 to 1.76). Furthermore, moder- ate/high stress was independently associated with poor 1-year health status, including a greater likelihood of angina, worse disease-specic and generic health status, and worse perceived health (p 0.01 for all). Conclusions Moderate/high perceived stress at the time of an AMI is associated with adverse long-term outcomes, even after adjustment for important confounding factors. Future studies need to examine whether stress mediates ob- served racial and socioeconomic disparities and whether novel interventions targeting chronic stress and coping skills can improve post-AMI outcomes. (J Am Coll Cardiol 2012;60:175663) 2012 by the American College of Cardiology Foundation Chronic stress is believed to contribute to both the devel- opment and progression of cardiovascular disease (1) and cardiovascular death (2,3). The mechanisms underlying this association are complex and include a combination of behavioral (e.g., smoking, lack of exercise, obesity [4], medication nonadherence) and physiological factors (e.g., increased blood pressure and pulse rates [5] reduced insulin sensitivity [6], increased platelet aggregation [7], endothe- lial dysfunction [8]). Although several studies have evalu- ated the adverse effect of stress on the cardiovascular system, many have focused on incident cardiac disease (13), re- stricted their analyses to select clinical populations (9,10), or have focused on particular types of stress, such as job strain (10,11), marital stress (10), and nancial strain (12). Few studies have evaluated the effect of chronic stress on indi- viduals recovering from an acute myocardial infarction (AMI), although studies have suggested that high job and marital strain can increase the risk of subsequent adverse cardiovascular events (10,11,13). Importantly, it is unknown whether high levels of chronic stress, assessed with a valid instrument that examines patient perception of stress and coping, are associated with adverse prognosis after an AMI. Given the From the Saint Lukes Mid America Heart Institute, Kansas City, Missouri. TRIUMPH was sponsored by a grant from the National Institutes of Health (National Heart, Lung, and Blood Institute): Washington University School of Medicine SCCOR grant P50HL077113-01. The funding organization did not play a role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; or preparation, review, or approval of the manuscript. Drs. Arnold and Smolderen are supported by an award from the American Heart Association Pharmaceutical Roundtable and David and Stevie Spina. Dr. Smolderen is also supported by the Netherlands Organization for Scientic Research [VENI grant no.: 916.11.179]; and received a research grant from W.L. Gore & Associates. Dr. Spertus owns the copyright to the Seattle Angina Questionnaire; received partial support from a Clinical and Translational Science Award (1UL1RR033179); received a research grant from Lilly; and is a consultant for United Healthcare. Drs. Buchanan and Li have reported that they have no relationships relevant to the contents of this paper to disclose. Manuscript received March 19, 2012; revised manuscript received June 6, 2012, accepted June 12, 2012. Journal of the American College of Cardiology Vol. 60, No. 18, 2012 2012 by the American College of Cardiology Foundation ISSN 0735-1097/$36.00 Published by Elsevier Inc. http://dx.doi.org/10.1016/j.jacc.2012.06.044 Downloaded From: http://content.onlinejacc.org/ on 01/08/2014 current focus of AMI management on optimizing prognosis, dening the prognostic import of perceived stress can illuminate an opportunity to assess and intervene in when risk-stratifying patients for subsequent treatment. Furthermore, the association between perceived stress and post-AMI angina and health status (function and quality of life[QoL]) outcomes has not been described. Although depressive symptoms have been shown to be associated with health status after adjusting for perceived stress in a cross- sectional analysis of patients with stable coronary artery disease, the independent association of stress with outcomes was not reported (14). Moreover, depressive symptoms are associated with both increased stress and poor health status and mortality. Therefore, exploring the association of stress with outcomes independent of depressive symptoms is important to disen- tangle these complex relationships and understand whether stress is an important factor to consider, beyond depression, when evaluating psychological distress in AMI patients and to identify novel potential areas for interventions to improve outcomes. To better clarify the association of chronic stress, assessed with the validated perceived stress scale (PSS) (15), on mortality and health status outcomes, we examined a large population of AMI patients in the multicenter TRIUMPH (Translational Research Investigating Underlying disparities in acute Myocardial infarction Patients Health status) study (16). The TRIUMPH study was an ideal population for this analysis, given its extensive collection of socioeconomic, clinical, and psychosocial factors that may confound the association between stress and outcomes. Methods Study population and protocol. Between June 2005 and December 2008, 4,340 patients from 24 U.S. hospitals were enrolled in the TRIUMPH study (16). Patients were required to have biomarker evidence of myocardial necrosis and additional clinical evidence supporting the diagnosis of an AMI, including prolonged ischemic signs/symptoms (20 min) or electrocardiographic ST-segment changes during the initial 24 h of admission. Baseline data were obtained through chart abstraction and a structured inter- view by trained research staff within 24 to 72 h after admission. Follow-up health status data were obtained 1 year after patients AMI with a standardized interview. Each participating hospital obtained institutional research board approval, and all patients provided written informed consent for baseline and follow-up assessments. Assessment of perceived stress. Patients levels of per- ceived stress were assessed at the time of AMI with the 4-item PSS (15). The PSS is a reliable and valid measure of subjects sense of control and condence in handling cir- cumstances over the past month. The 4 questions ask respondents to judge their ability to control important things, their condence in handling personal problems, a feeling of things going their way, and a feeling of being overwhelmed by difculties pil- ing up. Responses range from never (0 points) to very often (4 points), with the total PSS score ranging from 0 to 16 and higher scores indicating more perceived stress. Because the PSS is not a diagnostic instrument, there are no established cutoffs, only com- parisons within a sample. There- fore, we divided the scores into approximate quintiles and de- ned the highest quintile (PSS scores 9 to 16) as high stress in accordance with previous work (14) and the fourth quintile (PSS scores 6 to 8) as representing moderate stress. Outcomes assessment. Two- year all-cause mortality (available for 99% of patients) was deter- mined through a combination of follow-up interviews and a query of the Social Security Death Masterle. Health status was assessed using the Seattle Angina Questionnaire (SAQ) (17), the EuroQol Visual Analog Scale (VAS) (18), and the Medical Outcomes Study 12-Item Short Form (SF-12) (19). The SAQ is a reliable and valid 19-item questionnaire composed of 5 clinically important dimensions of health in patients with coronary artery disease. For this study, we focused on the SAQ angina frequency, physical limitations, and QoL scales. The scores for these domains range from 0 to 100, with higher scores indicating less disease burden. Based on previous work, SAQ angina frequency was cate- gorized as none (score of 100), monthly angina (score of 61 to 99), and weekly/daily angina (score of 0 to 60); SAQ physical limitations was categorized as slight to no limita- tion (score 75), mild limitation (score of 51 to 75), and moderate/severe limitation (score of 0 to 50); and SAQ QoL was categorized as excellent (score 75), good (score of 51 to 75), and fair/poor (score of 0 to 50) (20). The VAS is a measure of perceived general health that consists of a single-item, feeling thermometer, on which patients rate their general health state from 0 (worst imaginable) to 100 (best imaginable) (18). The SF-12 is a reliable and valid measure of generic health status (21) that provides summary component scales for overall physical (Physical Components Summary [PCS]) and mental (Men- tal Components Summary [MCS]) health using norm- based methods that standardize the scores to a mean of 50 and an SD of 10 (higher scores indicate better health status) (19). To facilitate the interpretation of our results, we dened poor health status as scores in the lowest quartile (VAS scores of 65; PCS scores of 35; MCS scores of 45). Statistical analysis. Baseline factors and long-term out- comes of the patients were compared across categories of stress using 1-way analysis of variance for continuous Abbreviations and Acronyms AMI acute myocardial infarction CI condence interval MCS Mental Components Summary OR odds ratio PCS Physical Components Summary PSS Perceived Stress Scale QoL quality of life SAQ Seattle Angina Questionnaire SF-12 Medical Outcomes Study 12-item Short Form VAS EuroQol Visual Analog Scale 1757 JACC Vol. 60, No. 18, 2012 Arnold et al. October 30, 2012:175663 Perceived Stress and AMI Outcomes Downloaded From: http://content.onlinejacc.org/ on 01/08/2014 variables and the chi-square test for categorical variables. A multivariable logistic regression model was used to deter- mine the factors associated with moderate/high versus low stress levels (PSS scores of 6 to 16 vs. scores of 0 to 5) among AMI patients. Variables included in the model were selected a priori and included demographics, socioeconomic status, comorbidities and clinical factors, and disease sever- ity indices. Cox proportional hazards regression models evaluated the impact of moderate/high stress versus low levels of stress on 2-year all-cause post-AMI mortality. Two sequential mor- tality models were constructed: adjusted for the GRACE (Global Registry of Acute Coronary Events) discharge score (22) to account for the clinical severity of AMI and fully adjusted with additional sociodemographic, clinical, and treatment factors. Multivariable logistic regression models assessed the independent association of moderate/high lev- els of stress with poor health status outcomes 1 year after AMI. Outcomes included SAQ angina frequency (none [score 100] vs. any [score 100]), SAQ physical limita- tions (slight to no limitation [score 75] vs. mild or greater limitation [score of 0-75]), SAQ QoL (excellent [score 75] vs. diminished [score of 0 to -75]), VAS (good [score 65] vs. poor [score 65]), SF-12 PCS (good [score 35] vs. poor [score 35]), and SF-12 MCS (good [score 45] vs. poor [score 45]). Covariates in all health status models and in the fully adjusted mortality model were selected a priori given their associations with stress and poor outcomes. These included age, sex, race, marital status, self-reported avoidance of care due to costs (a measure of economic barrier to medical care that has demonstrated strong association with poor long- term AMI outcomes [23]) hypertension, previous bypass surgery, diabetes, chronic lung disease, chronic heart failure, history of stroke or transient ischemic attack, body mass index, anemia, depressive symptoms (assessed with the 9-item Patient Health Questionnaire [24]), ST-segment elevation AMI, GRACE score (22), left ventricular func- tion (categorized as normal, mild, moderate, and severe dysfunction), in-hospital percutaneous coronary interven- tion, in-hospital bypass graft surgery, and the percentage of Joint Commission/Centers for Medicare and Medicaid Services quality measures received (of those for which the patient was eligible). The interaction between depressive symptoms and perceived stress was tested and found not to be signicant (p 0.05 for all analyses) and was not included in the nal models. Smoking status and low social support (assessed with the 7-item Enhancing Recovery in Coronary Heart Disease Social Support Inventory [25]) were chosen a priori not to be included in the models to avoid overadjusting because they were thought to be mark- ers or consequences of high stress rather than causes of it. Of note, these variables were strongly associated with higher stress levels but weakly associated with outcomes. Several sensitivity analyses were performed that included smoking, low social support, and baseline health status (to test the association between baseline stress levels and health status recovery) as additional covariates in the multivariable mod- els. All analyses were conducted using SAS version 9.2 (SAS Institute, Inc., Cary, North Carolina), and statistical signicance was determined by a 2-sided p value of 0.05. Missing data. Of the 4,340 patients enrolled in the TRI- UMPH study, 4,204 completed the PSS and were dis- charged alive. Baseline covariate data were fairly complete, with 12% of patients missing 1 covariate, 1% of patients missing 2 covariates, and an average missing rate of 0.14 items per patient. Missing baseline covariate data were imputed using multiple imputations with IVEware (Impu- tation and Variance Estimation Software; University of Michigans Survey Research Center, Institute for Social Research, Ann Arbor, Michigan). At 1 year, 279 partici- pants (6.6%) had died, 235 (5.6%) refused the interview, 78 (1.9%) did not complete the full interview, and 971 (23.1%) could not be contacted. Although unadjusted baseline PSS scores were higher for those alive but missing follow-up health status assessments (4.9 vs. 4.4, p 0.001), stress was not associated with missing follow-up in a multivariable logistic model (p 0.30). To further assess for potential bias in the health status outcomes assessments, we calcu- lated a nonparsimonious propensity score with successful follow-up as the dependent variable. An inversely weighted propensity score was assigned to each patient (26), and we then repeated our health status models to provide greater weight to the scores of patients who were most like those missing follow-up data. The results of this sensitivity analysis were consistent with the main analyses, and only the unweighted analyses are presented. Results Patient population. Of the 4,204 patients who were dis- charged alive after their AMI, 1,622 (38.6%) were consid- ered as having moderate or high levels of perceived stress in the 4 weeks preceding their AMI. The mean age of the patient population was 59 years, 67% were male, and 67% were Caucasian. Cardiac and noncardiac comorbidities were common, with 66% of patients having hypertension, 31% having diabetes, and 19% expressing depressive symptoms (Personal Health Questionnaire-9 score 10). Patients with higher stress levels before their AMI were younger and more likely to be female, unmarried, and non-Caucasian; have low social support, avoid care due to cost; and were less likely to have completed high school. Higher stress patients also had more hypertension, diabetes, smoking, depression, and obesity (Table 1). In terms of their clinical presentation, patients with higher stress levels were less likely to present with ST-segment elevation AMIs and had lower GRACE scores but had higher heart rates on admission and longer length of stays. Higher stress patients were less likely to undergo invasive management of their AMI, but guideline- recommended medical therapy at discharge was high and similar across categories of stress (Table 2). 1758 Arnold et al. JACC Vol. 60, No. 18, 2012 Perceived Stress and AMI Outcomes October 30, 2012:175663 Downloaded From: http://content.onlinejacc.org/ on 01/08/2014 Association between perceived stress and mortality outcomes. Among the 4,204 patients who survived the acute hospital- ization for their AMI, 432 (10.3%) died within the 2 years after hospital discharge. Patients with moderate/high perceived stress levels had increased mortality over the subsequent 2 years compared with those who reported low stress levels (12.9% vs. 8.6%; log-rank test, p 0.001) (Table 3, Fig. 1). In a Cox proportional hazards model adjusting for the GRACE discharge score, mod- erate/high stress levels were associated with an increased hazard of dying during the 2 years after AMI (hazard ratio:1.79; 95% condence interval [CI]: 1.48 to 2.16; p 0.001). After further adjusting for sociodemograph- ics, clinical factors including depressive symptoms, and treatment characteristics, moderate/high stress remained signicantly associated with increased 2-year mortality (hazard ratio: 1.42; 95% CI: 1.15 to 1.76; p 0.001) (Fig. 2). Association between perceived stress and health status outcomes. Moderate/high perceived stress levels were as- sociated with worse 1-year health status outcomes, includ- ing increased angina, more physical limitations due to angina, worse disease-specic QoL, and worse physical and mental health status (Table 3). In multivariable logistic regression models adjusting for sociodemographic and clin- ical factors including depressive symptoms, patients with moderate/high stress during the month preceding their AMI had a 1.41 increased odds of having angina 1 year after their AMI (95% CI: 1.14 to 1.75; p 0.001) (Fig. 2) and a 1.41 increased odds of having a diminished disease- specic QoL (95% CI: 1.13 to 1.77) (p 0.003). Patients with moderate/high stress were also more likely to report poor perceived health, as assessed with the VAS (odds ratio [OR]: 1.79; 95% CI: 1.45 to 2.20; p 0.001), SF-12 physical health status (OR: 1.43; 95% CI: 1.14 to 1.78; p 0.002), and SF-12 mental health status (OR: 2.34; 95% CI: 1.87 to 2.93; p 0.001). Perceived stress was not signi- cantly associated with angina-related physical limitations after multivariable adjustment (p 0.11) (Fig. 2, Online Tables 1a and 1b). Sensitivity analyses. To support the robustness of our ndings, we performed a number of sensitivity analyses. Although the PSS has a 1-month recall period and would be expected to capture patients perceived stress before their MI, it is possible that the acute stress of an AMI might alter the intended recall frame of the instrument. To address this possibility, we repeated our analyses using a 1-month post-AMI PSS assessment. Of the patients, 71% reported the same level of perceived stress at baseline and 1 month after their AMI (66% of those with moderate/high stress at baseline and 73% of those with low stress at baseline). Moderate/high stress at the 1-month follow-up was asso- ciated with an increased hazard of 2-year mortality (unad- justed HR: 1.41; 95% CI: 1.0 to 1.87; p 0.016) and increased odds of poor 1-year health status across all the health status domains (Online Table 2). Second, although we explicitly excluded smoking and social support from the multivariable models to avoid overadjustment, we per- Baseline Characteristics of Patients With Various Levels of Chronic Perceived Stress Table 1 Baseline Characteristics of Patients With Various Levels of Chronic Perceived Stress Perceived Stress Level p Value Low (n 2,582) Moderate (n 1,027) High (n 595) Sociodemographics Age, yrs 60.5 12.3 57.6 11.9 54.2 11.0 0.001 Male 70.4 64.2 57.0 0.001 Caucasian race 69.8 63.5 63.7 0.001 Married 56.8 49.8 38.9 0.001 Low social support 9.4 24.1 39.3 0.001 Lives alone 23.2 25.4 28.4 0.024 High school education 82.0 75.0 74.7 0.001 Avoids care due to cost 18.4 30.7 48.6 0.001 Comorbidities Hypertension 64.2 69.4 71.4 0.001 Previous bypass graft surgery 11.4 11.8 10.8 0.822 Diabetes mellitus 27.5 34.8 36.5 0.001 Current smoking 34.8 44.1 51.7 0.001 Anemia on admission 19.7 23.4 25.6 0.001 Chronic lung disease 6.7 7.2 9.7 0.032 History of heart failure 7.1 9.3 12.9 0.001 GFR 60 ml/min 24.4 24.4 25.5 0.843 Body mass index, kg/m 2 29.2 6.2 29.8 6.8 30.5 6.8 0.001 Stroke or TIA 6.7 7.1 8.1 0.492 Depressive symptoms 6.9 24.4 60.6 0.001 Values are mean SD or %. GFR glomerular ltration rate; TIA transient ischemic attack. 1759 JACC Vol. 60, No. 18, 2012 Arnold et al. October 30, 2012:175663 Perceived Stress and AMI Outcomes Downloaded From: http://content.onlinejacc.org/ on 01/08/2014 formed sensitivity analyses including these 2 variables. The association between stress and outcomes was not materially affected by the addition of low social support or smoking to the multivariable models (Online Table 3). Third, although we were specically interested in whether chronic stress is associated with adverse health status outcomes after an AMI, we additionally investigated whether stress is associ- ated with poor health status recovery by additionally adjust- ing for baseline health status in the multivariable models. Although the relationship was attenuated, moderate/high stress at baseline remained signicantly associated with poorer health status recovery 1 year after their AMI (Online Table 4). Fourth, because depressive symptoms and stress are often coexisting, we compared the outcomes of patients with low versus moderate/high stress among patients with and without depressive symptoms. Within each category of depressive symptoms, patients with moderate/high stress were more likely to die and report worse angina and QoL compared with patients reporting low stress (Online Table 5). Finally, because deaths were not adjudicated, we are only able to report all-cause mortality and not cardiac- specic mortality. However, moderate/high stress was asso- AMI Presentation and Treatment Characteristics of Patients With Various Levels of Chronic Perceived Stress Table 2 AMI Presentation and Treatment Characteristics of Patients With Various Levels of Chronic Perceived Stress Perceived Stress Level p Value Low (n 2,582) Moderate (n 1,027) High (n 595) Clinical presentation ST-segment elevations 45.2 39.5 39.0 0.001 Peak troponin 29.2 76.3 26.5 64.8 28.0 72.2 0.607 GRACE risk score 102.4 29.9 98.9 30.2 93.3 28.7 0.001 Left ventricular systolic function 0.158 Normal 61.5 64.0 60.2 Mild dysfunction 20.6 17.2 18.3 Moderate dysfunction 10.7 11.0 12.8 Severe dysfunction 7.2 7.8 8.7 Initial heart rate 81.9 21.8 82.7 21.7 85.3 24.0 0.003 Initial systolic blood pressure 142.9 30.3 143.5 29.3 143.4 31.5 0.813 hsCRP levels 3.8 5.1 3.5 4.7 3.5 4.4 0.471 Length of stay 5.3 5.7 5.8 7.4 5.9 6.5 0.02 Hospital procedures Cardiac catheterization 93.6 89.2 90.6 0.001 PCI 67.2 60.1 63.5 0.001 Bypass graft surgery 9.5 9.3 9.1 0.941 Medications at discharge Aspirin 94.9 93.1 94.1 0.107 Beta-blocker 90.8 90.0 88.3 0.165 ACE inhibitor or ARB 74.2 73.3 76.4 0.372 Statin 88.5, 86.5 88.6 0.226 % of QOC indicators 90.6 14.8 89.0 16.3 88.9 16.1 0.004 Values are % or mean SD. ACE angiotensin-converting enzyme; AMI acute myocardial infarction; ARB angiotensin receptor blocker; GRACE Global Registry of Acute Coronary Events; hsCRP high-sensitivity C-reactive protein; PCI percutaneous coronary intervention; QOC quality of care. Long-Term Post-AMI Outcomes of Patients With Various Levels of Chronic Perceived Stress Table 3 Long-Term Post-AMI Outcomes of Patients With Various Levels of Chronic Perceived Stress Perceived Stress Level p Value Low Moderate High Mortality rate, 2 year 8.6 13.0 12.8 0.001 Health status outcomes, 1 year SAQ angina frequency 94.6 13.9 91.6 16.7 87.0 22.1 0.001 SAQ physical limitations 94.2 15.2 91.6 19.0 82.8 26.3 0.001 SAQ quality of life 85.2 18.6 79.3 20.9 68.6 26.5 0.001 EuroQol Visual Analog Scale 78.7 18.5 71.9 20.4 63.1 24.4 0.001 SF-12 physical health 45.1 11.7 42.0 11.9 38.6 12.7 0.001 SF-12 mental health 54.4 8.8 49.9 10.0 43.2 12.6 0.001 Values are % or mean SD. AMI acute myocardial infarction; SAQ Seattle Angina Questionnaire; SF-12 Medical Outcomes Study 12-Item Short Form. 1760 Arnold et al. JACC Vol. 60, No. 18, 2012 Perceived Stress and AMI Outcomes October 30, 2012:175663 Downloaded From: http://content.onlinejacc.org/ on 01/08/2014 ciated with an increased hazard of cardiac rehospitalizations. There was no signicant difference between the rate of coronary revascularization over the year of follow-up be- tween those with moderate/high and low stress levels (Online Fig. 1). Discussion In a large contemporary registry of AMI patients, moderate to high perceived stress over the month before an AMI was associated with increased long-term mortality, despite the fact that patients with increased perceived stress levels were younger and less likely to have ST-segment elevation AMIs and had lower GRACE scores, all of which are generally associated with lower mortality. Although high perceived stress levels were associated with depressive symptoms, the association between increased stress and increased mortality persisted after adjustment for multiple factors, including depressive symptoms. Thus, chronic stress outside the context of depressive symptoms is independently associated with adverse outcomes. Furthermore, increased stress levels before AMI were associated with worse health status outcomes and poorer health status recovery after AMI, including more angina and worse QoL. Although the mechanisms of these associations remain unknown, these ndings suggest that patients perceived stress is an impor- tant factor to consider when evaluating patients risk of poor long-term outcomes. In addition, many covariates that are associated with increased perceived stress have been shown in previous studies to be associated with poor outcomes post-AMI (e.g., female sex [(27] low social support [27], low socioeconomic status [23], smoking [28]), suggesting that some of the effect of these factors on outcomes may be mediated by poorer coping skills and increased stress. Previous studies. Our ndings substantially extend the previous literature surrounding the association between chronic stress and cardiovascular outcomes. Among work- ing people, several studies have demonstrated that increased job strain is associated with the development and progres- sion of cardiovascular disease (13,29) and the risk of recurrent adverse cardiovascular events after an AMI (9,11,29). However, this nding has not been consistent (30,31), particularly among women (10). Other studies have shown high marital (10) and nancial (12) strain to be associated with an increased risk of recurrent adverse events after AMI among women. Our study extends the evaluation of stress and prognosis in a large, diverse, unselected AMI population that included patients regardless of employment, marital status, or age, further supporting the generalizability of our ndings. Importantly, we were able to adjust for multiple potential confounders of the relationship between stress and mortality, including depressive symptoms and socioeconomic status, further supporting an independent relationship between perceived stress and outcomes. Second, we have extended the literature by broadening our outcomes to include health status and QoL. Limited evidence is available demonstrating that mental stress may be related to anginal symptoms (32,33), and the cross- sectional Heart and Soul Study of 1,024 U.S. patients with stable coronary artery disease suggested that increased per- ceived stress was independently associated with greater physical limitations from angina and worse disease-specic and generic health status. However, our study is the rst to prospectively document the independent association between perceived 0.5 1 2 4 SAQ Angina Frequency SAQ Physical Limitations SAQ Quality of Life SF-12 Physical Health SF-12 Mental Health 1.42 (1.15, 1.76) 1.41 (1.14, 1.75) 1.30 (0.94, 1.79) 1.41 (1.13, 1.77) 1.79 (1.45, 2.20) 1.43 (1.14, 1.78) 2.34 (1.87, 2.93) Mortality EuroQol Visual Analog Scale Disease-Specific Health Status Generic Health Status Figure 2 Association Between Moderate/High and Low Perceived Stress Levels and Long-Term Outcomes Multivariable-adjusted hazard/odds ratios of the association between moder- ate/high perceived stress levels (stress scores of 6 to 16) and low stress lev- els (stress scores of 0 to 5) and mortality and health status outcomes. Models were adjusted for age, sex, race, marital status, self-reported avoidance of care due to costs, hypertension, previous bypass surgery, diabetes, chronic lung disease, chronic heart failure, history of stroke or transient ischemic attack, body mass index, anemia, depressive symptoms, ST-segment eleva- tions on admission, left ventricular function, Global Registry of Acute Coronary Events score (22), in-hospital bypass surgery, in-hospital percutaneous coro- nary intervention, and the percentage of quality of care indicators received. SAQ Seattle Angina Questionnaire; SF-12 Medical Outcomes Study 12-Item Short Form. Months S u r v i v a l
( % ) 0 3 6 9 12 15 18 21 24 80 90 100 Low stress Moderate/high stress Log-rank p=0.0001 Figure 1 Kaplan-Meier Survival Curves for Moderate/High Versus Low Perceived Stress Levels Two-year survival after acute myocardial infarction in patients with moderate/ high perceived stress levels (stress scores of 6 to 16) versus low stress levels (stress scores of 0 to 5). 1761 JACC Vol. 60, No. 18, 2012 Arnold et al. October 30, 2012:175663 Perceived Stress and AMI Outcomes Downloaded From: http://content.onlinejacc.org/ on 01/08/2014 stress and a broad range of validated disease-specic and generic health status outcomes after an AMI. Potential mechanisms. The mechanisms underlying the association between chronic perceived stress and poor out- comes after an AMI are complex and likely include a combination of behavioral (4) and physiological (58) factors. From a physiological standpoint, acute mental stress can have adverse consequences on the cardiovascular system in terms of inducing myocardial ischemia. Structured men- tal stress tests can provoke ischemia of similar or greater severity and extent to exercise in 40% to 70% of patients with coronary artery disease (34,35). Furthermore, myocar- dial ischemia induced by mental stress predicts ischemia during daily life and increases the risk of fatal and nonfatal cardiac events (36,37). These ndings suggest that post- AMI patients who have high levels of mental stress may experience adverse physiological effects that may, in part, explain their poor outcomes. Furthermore, behavioral fac- tors that are both associated with high stress and poor outcomes, such as medication nonadherence and nonpar- ticipation in cardiac rehabilitation, should be explored as potential mediators of the effect of stress on poor outcomes. Perceived stress and other psychosocial distress. Previous considerations of the psychosocial risks associated with poor post-AMI outcomes have primarily focused on recognition and treatment of depressive symptoms. Although stress is associated with other manifestations of psychological dis- tress, our study demonstrates that an increased chronic perceived stress level at the time of an AMI is independently associated with increased mortality and worse health status after the AMI, even after accounting for depressive symp- toms. Thus, signicant stress that occurs outside the context of depression may also independently and adversely portend poor outcomes. Furthermore, there has been recent atten- tion given to the adverse impact of other psychosocial factors, such as anger, anxiety, and social support on mortality (3840) and health status (41) in patients with coronary artery disease. Although the psychological con- cepts of stress, anxiety, depression, anger, and hostility are not interchangeable, there is overlap among these concepts, and it is unclear how much of the PSS captures psycholog- ical distress due to these factors that are related to but distinct from chronic stress. Nevertheless, although the PSS may not be able to disentangle these psychological mecha- nisms, it is a simple, 4-item instrument that can be easily implemented in clinical care to identify high-risk patients with psychological distress that may benet from a multi- modal psychological evaluation and intervention. Expand- ing psychosocial interventions to address factors such as chronic stress, coping skills, anxiety, and other psychosocial impairments may improve post-AMI outcomes in vulnera- ble patients. Such multicomponent approaches have shown success on a small scale. For example, in a study of 237 Swedish women with coronary disease, patients randomized to a group-based psychosocial intervention designed to improve coping skills, reduce stress, and improve social support had signicantly reduced mortality compared with patients who received usual care (42). Study limitations. Our ndings should be interpreted in the context of the following potential limitations. First, although the assessment of perceived stress at the time of the AMI was done with the PSS, a psychometrically appropriate instrument with a recall period of 1 month, it is unclear whether the acute stress of the AMI affected patients recollection of their pre-AMI stress levels. However, when we repeated the anal- yses using the 1-month follow-up PSS scores, the results were consistent (and across most domains, the associations were even stronger), which increases our condence in the prognos- tic association of perceived stress with post-AMI outcomes. Second, the PSS is not a diagnostic instrument, and thus there are no established score cutoffs. Instead, the authors of the instrument recommend using cutoffs based on the study sample. We used quintiles because the uppermost quintile in our study sample (PSS scores of 9 to 16) correlated with a threshold for high perceived stress used in the Heart and Soul Study (14). Although the use of different score cutoffs may have altered the point estimates of our ndings, we found a graded, nearly continuous relationship across quintiles of PSS scores. Third, as patients missing follow-up information re- ported increased perceived stress and worse health status at baseline, the loss to follow-up may have biased the health status/QoL analyses, although our propensity-weighted anal- yses did not suggest a signicant bias from loss to follow-up. Finally, given the interplay of multiple psychosocial and socio- economic aspects with each other (including smoking, anxiety, depressive symptoms, and socioeconomic status), it is difcult to disentangle and isolate the specic relationship between stress and outcomes. Properly teasing apart the potentially causative pathway would require formal mediation or path analyses with prospective, serial measurements of stress, which are important areas for further research. Conclusions We found that moderate to high levels of perceived stress over the month before an AMI were associated with increased long-term post-AMI mortality, despite the fact that patients with increased perceived stress levels were younger and less ill. Furthermore, increased stress was associated with poor long-term health status outcomes. These ndings suggest that it is important to consider patients perceived stress levels when evaluating their risk for poor long-term outcomes. Although likely resulting from a combination of multiple biological and behavioral factors, further exploration is needed to determine the mechanisms of this increased hazard. In addition, it will be important to examine whether interventions targeting chronic stress levels and coping skills in AMI populations may attenuate the increased morbidity and mortality risk from increased stress. 1762 Arnold et al. JACC Vol. 60, No. 18, 2012 Perceived Stress and AMI Outcomes October 30, 2012:175663 Downloaded From: http://content.onlinejacc.org/ on 01/08/2014 Reprint requests and correspondence: Dr. Suzanne V. Arnold, St. Lukes Mid America Heart Institute, 4401 Wornall Road, Kansas City, Missouri 64111. E-mail: suz.v.arnold@gmail.com. REFERENCES 1. Rosengren A, Hawken S, Ounpuu S, et al. Association of psychosocial risk factors with risk of acute myocardial infarction in 11119 cases and 13648 controls from 52 countries (the INTERHEART study): case- control study. Lancet 2004;364:95362. 2. Iso H, Date C, Yamamoto A, et al. Perceived mental stress and mortality from cardiovascular disease among Japanese men and wom- en: the Japan Collaborative Cohort Study for Evaluation of Cancer Risk Sponsored by Monbusho (JACC Study). Circulation 2002;106: 122936. 3. Matthews KA, Gump BB. Chronic work stress and marital dissolution increase risk of posttrial mortality in men from the Multiple Risk Factor Intervention Trial. Arch Intern Med 2002;162:30915. 4. Rod NH, Gronbaek M, Schnohr P, Prescott E, Kristensen TS. Perceived stress as a risk factor for changes in health behaviour and cardiac risk prole: a longitudinal study. J Intern Med 2009;266:46775. 5. Steptoe A, Roy MP, Evans O, Snashall D. Cardiovascular stress reactivity and job strain as determinants of ambulatory blood pressure at work. J Hypertens 1995;13:20110. 6. Moberg E, Kollind M, Lins PE, Adamson U. Acute mental stress impairs insulin sensitivity in IDDM patients. Diabetologia 1994;37: 24751. 7. Jain S, Mills PJ, von Kanel R, Hong S, Dimsdale JE. Effects of perceived stress and uplifts on inammation and coagulability. Psy- chophysiology 2007;44:15460. 8. Ghiadoni L, Donald AE, Cropley M, et al. Mental stress induces transient endothelial dysfunction in humans. Circulation 2000;102: 24738. 9. Theorell T, Perski A, Orth-Gomer K, Hamsten A, de Faire U. The effects of the strain of returning to work on the risk of cardiac death after a rst myocardial infarction before the age of 45. Int J Cardiol 1991;30:617. 10. Orth-Gomer K, Wamala SP, Horsten M, Schenck-Gustafsson K, Schneiderman N, Mittleman MA. Marital stress worsens prognosis in women with coronary heart disease: the Stockholm Female Coronary Risk Study. JAMA 2000;284:300814. 11. Aboa-Eboule C, Brisson C, Maunsell E, et al. Job strain and risk of acute recurrent coronary heart disease events. JAMA 2007;298:165260. 12. Georgiades A, Janszky I, Blom M, Laszlo KD, Ahnve S. Financial strain predicts recurrent events among women with coronary artery disease. Int J Cardiol 2009;135:17583. 13. Ruberman W, Weinblatt E, Goldberg JD, Chaudhary BS. Psychos- ocial inuences on mortality after myocardial infarction. N Engl J Med 1984;311:5529. 14. Ruo B, Rumsfeld JS, Hlatky MA, Liu H, Browner WS, Whooley MA. Depressive symptoms and health-related quality of life: the Heart and Soul Study. JAMA 2003;290:21521. 15. Cohen S, Kamarck T, Mermelstein R. A global measure of psycho- logical stress. J Health Soc Behav 1983;24:38596. 16. Arnold SV, Chan PS, Jones PG, et al. Translational Research Investigating Underlying Disparities in Acute Myocardial Infarction Patients Health Status (TRIUMPH): design and rationale of a prospective multicenter registry. Circ Cardiovasc Qual Outcomes 2011;4:46776. 17. Spertus JA, Winder JA, Dewhurst TA, et al. Development and evaluation of the Seattle Angina Questionnaire: a new functional status measure for coronary artery disease. J Am Coll Cardiol 1995;25:33341. 18. EuroQol--a new facility for the measurement of health-related quality of life. The EuroQol Group. Health Policy 1990;16:199208. 19. Ware J Jr., Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care 1996;34:22033. 20. Spertus JA, Salisbury AC, Jones PG, Conaway DG, Thompson RC. Predictors of quality-of-life benet after percutaneous coronary inter- vention. Circulation 2004;110:378994. 21. Muller-Nordhorn J, Roll S, Willich SN. Comparison of the short form (SF)-12 health status instrument with the SF-36 in patients with coronary heart disease. Heart 2004;90:5237. 22. Eagle KA, Lim MJ, Dabbous OH, et al. A validated prediction model for all forms of acute coronary syndrome: estimating the risk of 6-month postdischarge death in an international registry. JAMA 2004;291:272733. 23. Rahimi AR, Spertus JA, Reid KJ, Bernheim SM, Krumholz HM. Financial barriers to health care and outcomes after acute myocardial infarction. JAMA 2007;297:106372. 24. Kroenke K, Spitzer RL, Williams JB. The PHQ-9: validity of a brief depression severity measure. J Gen Intern Med 2001;16:60613. 25. Vaglio J Jr., Conard M, Poston WS, et al. Testing the performance of the ENRICHD Social Support Instrument in cardiac patients. Health Qual Life Outcomes 2004;2:24. 26. Lunceford JK, Davidian M. Stratication and weighting via the propensity score in estimation of causal treatment effects: a compara- tive study. Stat Med 2004;23:293760. 27. Norris CM, Spertus JA, Jensen L, Johnson J, Hegadoren KM, Ghali WA. Sex and gender discrepancies in health-related quality of life outcomes among patients with established coronary artery disease. Circ Cardiovasc Qual Outcomes 2008;1:12330. 28. Maddox TM, Reid KJ, Spertus JA, et al. Angina at 1 year after myocardial infarction: prevalence and associated ndings. Arch Intern Med 2008;168:13106. 29. Schnall PL, Landsbergis PA, Baker D. Job strain and cardiovascular disease. Annu Rev Public Health 1994;15:381411. 30. Eaker ED, Sullivan LM, Kelly-Hayes M, DAgostino RB Sr., Ben- jamin EJ. Does job strain increase the risk for coronary heart disease or death in men and women? The Framingham Offspring Study. Am J Epidemiol 2004;159:9508. 31. Greenlund KJ, Kiefe CI, Giles WH, Liu K. Associations of job strain and occupation with subclinical atherosclerosis: the CARDIA Study. Ann Epidemiol 2010;20:32331. 32. Verthein U, Kohler T. The correlation between everyday stress and angina pectoris: a longitudinal study. J Psychosom Res 1997;43:2415. 33. Freedman SB, Wong CK. Triggers of daily life ischaemia. Heart 1998;80:48992. 34. Gullette EC, Blumenthal JA, Babyak M, et al. Effects of mental stress on myocardial ischemia during daily life. JAMA 1997;277:15216. 35. Ramachandruni S, Fillingim RB, McGorray SP, et al. Mental stress provokes ischemia in coronary artery disease subjects without exercise- or adenosine-induced ischemia. J Am Coll Cardiol 2006;47:98791. 36. Sheps DS, McMahon RP, Becker L, et al. Mental stress-induced ischemia and all-cause mortality in patients with coronary artery disease: results from the Psychophysiological Investigations of Myo- cardial Ischemia study. Circulation 2002;105:17804. 37. Jiang W, Babyak M, Krantz DS, et al. Mental stress-induced myo- cardial ischemia and cardiac events. JAMA 1996;275:16516. 38. Chida Y, Steptoe A. The association of anger and hostility with future coronary heart disease: a meta-analytic review of prospective evidence. J Am Coll Cardiol 2009;53:93646. 39. Roest AM, Martens EJ, Denollet J, de Jonge P. Prognostic association of anxiety post myocardial infarction with mortality and new cardiac events: a meta-analysis. Psychosom Med 2010;72:5639. 40. Leifheit-Limson EC, Reid KJ, Kasl SV, et al. The role of social support in health status and depressive symptoms after acute myocar- dial infarction: evidence for a stronger relationship among women. Circ Cardiovasc Qual Outcomes 2010;3:14350. 41. Arnold SV, Spertus JA, Ciechanowski PS, et al. Psychosocial modulators of angina response to myocardial ischemia. Circulation 2009;120:12633. 42. Orth-Gomer K, Schneiderman N, Wang HX, Walldin C, Blom M, Jernberg T. Stress reduction prolongs life in women with coronary disease: the Stockholm Womens Intervention Trial for Coronary Heart Disease (SWITCHD). Circ Cardiovasc Qual Outcomes 2009;2:2532. Key Words: health status y mortality y myocardial infarction y stress. APPENDIX For supplemental tables and a gure, please see the online version of this article. 1763 JACC Vol. 60, No. 18, 2012 Arnold et al. October 30, 2012:175663 Perceived Stress and AMI Outcomes Downloaded From: http://content.onlinejacc.org/ on 01/08/2014
Relationship Between Disease Duration and Preventive Motivation With Quality of Life of Patients With Coronary Heart Disease at Productive Age in Cardiology
Illness Beliefs Among Patients With Chronic Widespread Pain - Associations With Self-Reported Health Status, Anxiety and Depressive Symptoms and Impact of Pain